segformer_cracks_v2

Ventalik/segformer_cracks_v2

segformer_cracks_v2 is a machine learning model from Ventalik. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as other.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

2% of all-time downloads

All-time downloads

645

Public

Repo size

756 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin189 MB · 100%

At a glance

Library: transformers
License: other
Model type: segformer
Access: Public
Created: Oct 29, 2023
Updated: Mar 23, 2024
SHA: fe7d4a59

Base models

nvidia/mit-b3

Model card

From the Hugging Face model README

segformer_cracks_v2

This model is a fine-tuned version of nvidia/mit-b3 on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.0487
Mean Iou: 0.7770
Mean Accuracy: 0.8378
Overall Accuracy: 0.9803
Per Category Iou: [0.9797338315128605, 0.5741766454919031]
Per Category Accuracy: [0.9922972168222592, 0.683379436844461]

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 3e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Per Category Iou	Per Category Accuracy
0.0651	1.0	1541	0.0557	0.7543	0.8089	0.9785	[0.9779508845325062, 0.5306877773356943]	[0.9928050418870356, 0.6249652426334233]
0.0491	2.0	3082	0.0522	0.7629	0.8195	0.9792	[0.9786365168660656, 0.547127323866261]	[0.9926515789371488, 0.6463984572343288]
0.047	3.0	4623	0.0546	0.7519	0.7995	0.9787	[0.9782301112609909, 0.5255903669562288]	[0.9938779990669149, 0.6050380673194482]
0.0456	4.0	6164	0.0510	0.7671	0.8248	0.9795	[0.9789779157265575, 0.5552760281623849]	[0.9925739681553413, 0.6570897409273347]
0.0442	5.0	7705	0.0509	0.7714	0.8373	0.9794	[0.978875460970978, 0.5639303830124743]	[0.9914357800230804, 0.6831791371245445]
0.0439	6.0	9246	0.0502	0.7701	0.8254	0.9799	[0.9794166328395213, 0.5607787508694683]	[0.9929877378064772, 0.6578722690116804]
0.0433	7.0	10787	0.0500	0.7738	0.8361	0.9799	[0.979329611841869, 0.5681732097856239]	[0.9920131586953994, 0.6802192080134767]
0.0427	8.0	12328	0.0499	0.7775	0.8406	0.9802	[0.9796585654955907, 0.575368007293997]	[0.9919935734749642, 0.6891110848654891]
0.0424	9.0	13869	0.0495	0.7780	0.8411	0.9802	[0.9797042588379468, 0.5763456606457494]	[0.9919931330972266, 0.6902882749336847]
0.042	10.0	15410	0.0487	0.7770	0.8378	0.9803	[0.9797338315128605, 0.5741766454919031]	[0.9922972168222592, 0.683379436844461]

Framework versions

Transformers 4.33.1
Pytorch 2.0.1
Datasets 2.14.5
Tokenizers 0.13.3

Library: transformers
Type: segformer
License: other
Created: Oct 29, 2023
Updated: Mar 23, 2024